Improved Optimization Scheme based on Rule for Energy Efficiency and Multiple Users Comfort Environment in Smart Home

Abstract

Smart homes and residential buildings are becoming one of the interesting and challenging research topics in order to satisfy what occupants' need in the certain building environment. At the same time, the total amount of energy consumption in smart home and building environment has been increasing rapidly since last few years. Therefore, many scientific researchers have been given huge attention to the energy control and management in smart home and residential building environment. Several proposals based on optimization algorithms and other technologies exist in literature that has been tried to solve the challenge between energy consumption and occupant's comfort index. In this thesis, we proposed rule based optimization scheme for reducing power consumption, an optimization scheme based on dynamic user setting for multi-users, and optimization scheme based on prediction of indoor environment parameters in order to increase user comfort index and consume less energy in the smart home area. Previously, we have already implemented optimization algorithms such as Ant colony optimization and Incremental genetic algorithm in order to increase user satisfaction level and energy efficiency. The energy control system using algorithms aimed to find highest optimal set points and increase the occupant's overall satisfaction in building environment. It gives a high overall comfort index and less power consumption results. However, there are still ways to get higher comfort index results with less energy consumption. Therefore the purpose of this thesis is aimed to increase occupant's comfort index and consume less power through optimal environment set points which is using rule based optimization. In addition, we considered multi user set point setting for every members of the home. Thus every user is able to customize their comfort condition ranges by dynamic user set point settings for multi-users. The third proposed idea is about predicting indoor environment parameters to consume less power. As a result, RBO reduced power consumption by 24.32% as compared to GA, 10.26% as compared to IGA, and 25.72% as compared to ACO. To satisfy multi users' comfort in smart home, we proposed dynamic user set points setting by three methods. Among the three methods, max-min based user set point setting consumed highest power. Average based user set point setting reduced power by 4.28% as compared to max-min based user set point setting and min-max based user set point setting reduced power by 8.74% as compared to max-min based user set point setting. Finally, we compared predicted indoor environment parameters and unpredicted indoor parameters. For illumination and air quality control, the results were almost similar. But, the prediction of indoor parameters, for temperature control, ABS, and RBO based system reduced power consumption by 2% as compared to unpredicted indoor parameters, ABS, and RBO based system. Prediction of indoor parameters, for temperature control Max-min, and RBO based system reduced power consumption by 0.71% as compared to unpredicted indoor parameters, Max-min, and RBO based system. Similarly, prediction of indoor parameters, for temperature control Min-max, and RBO based system reduced power consumption by 3.28% as compared to unpredicted indoor parameters, Min-max, and RBO based system.